Go SDK and examples for MiniMax APIs.

• Speech APIs
  • synchronous HTTP TTS
  • streaming TTS
  • official WebSocket TTS
  • async TTS task submit/query
• File upload API
• Voice APIs
  • list voices
  • voice design
  • voice clone
  • voice delete
  • typed clone/prompt audio uploads
• Image APIs
  • text-to-image generation
  • image-to-image generation
• Music APIs
  • lyrics generation and editing
  • non-streaming music generation
  • cover preprocess and cover generation
• Video APIs
  • text-to-video task submit/query
  • image-to-video task submit
  • first-last-frame video task submit
  • subject-reference video task submit

The detailed API inventory lives in docs/. Current coverage is:

Implemented:

• File upload: File.Upload supports multipart upload and normalized upload metadata.
• File list: File.List lists stored files by MiniMax purpose.
• File retrieve: File.Retrieve retrieves normalized metadata for a generated or uploaded file.
• File download: File.Download opens a raw file content stream for generated files.
• File delete: File.Delete deletes a stored file by file ID and purpose.
• Voice list: Voice.ListVoices queries available system, cloned, and generated voices.
• Voice design: Voice.DesignVoice creates a custom voice from a prompt and preview text.
• Voice delete: Voice.DeleteVoice deletes an owned generated/cloned voice by ID.
• Voice clone audio uploads: Voice.UploadCloneAudio and Voice.UploadPromptAudio wrap official file upload purposes.
• Speech T2A HTTP: Speech.Synthesize supports synchronous HTTP TTS with stable official audio, language, subtitle, and output fields.
• Speech T2A WebSocket: Speech.OpenWebSocket implements the official /ws/v1/t2a_v2 task protocol separately from HTTP stream.
• Speech T2A async: SpeechAsync.SubmitAsync and SpeechAsync.GetAsyncTask support stable official create/query fields and metadata normalization.
• Image T2I: Image.GenerateTextToImage generates images from text prompts.
• Image I2I: Image.GenerateImageToImage generates images from prompts and subject references.
• Music lyrics generation: Music.GenerateLyrics writes or edits lyrics for music workflows.
• Music generation: Music.Generate creates non-streaming songs, instrumental tracks, and cover music.
• Music cover preprocess: Music.PreprocessCover extracts cover features and formatted lyrics for two-step cover workflows.
• Video T2V create: Video.CreateTextToVideo creates async text-to-video tasks.
• Video I2V create: Video.CreateImageToVideo creates async image-to-video tasks.
• Video FL2V create: Video.CreateFirstLastFrameVideo creates async first-last-frame video tasks.
• Video S2V create: Video.CreateSubjectReferenceVideo creates async subject-reference video tasks.
• Video generation query: Video.GetTask queries async video task status and generated file IDs.

Partially implemented:

• Speech T2A streaming: Speech.OpenStream remains an HTTP/SSE helper for source compatibility; use Speech.OpenWebSocket for the official WebSocket protocol.
• Voice clone: Voice.CloneVoice supports audio_url and file_id; full official preview/prompt clone fields can be added separately.
• Music streaming: Music.Generate rejects stream=true; official streaming music response handling is not implemented yet.

Planned:

• Remaining video generation APIs: video agent tasks.
• Text and model APIs: OpenAI/Anthropic-compatible chat, Responses, token estimation, and model list/retrieve endpoints.

• Go 1.26+
• MiniMax API key

Set your API key:

export MINIMAX_API_KEY="your_api_key"

Check runnable examples:

go run ./examples/speech -h
go run ./examples/speech async -h
go run ./examples/speech stream -h
go run ./examples/speech websocket -h
go run ./examples/speech http -h
go run ./examples/voice/list -h
go run ./examples/voice/clone -h
go run ./examples/voice/upload -h
go run ./examples/voice/delete -h
go run ./examples/file -h
go run ./examples/image -h
go run ./examples/music -h
go run ./examples/video -h

go fmt ./...
go build ./...
go vet ./...
go test ./...

• client.go: SDK client and service wiring
• speech*.go: speech sync/stream/async APIs
• voice.go: voice-related APIs
• image.go: image generation APIs
• music.go: music generation and lyrics APIs
• video.go: video generation task APIs
• file.go: file management APIs
• docs/: official API inventory and implementation status by interface
• internal/: transport/protocol/stream/codec internals
• examples/: runnable CLI demos